Chomski

Infobox programming language
name = chomski
paradigm = scripting language
year = 2007
designer = mj bishop
typing = none; all data is treated as a string
implementations = [http://bumble.sourceforge.net/machine/c/ chomski]
influenced_by = Sed, Awk
operating_system = Cross-platform
website = [http://bumble.sourceforge.net/machine/ bumble.sourceforge.net/machine/]

chomski (named after the noted linguist Noam Chomsky) refers to a command line utility and languagewhich is can be used to parse and transform text patterns. Chomski (a) parses text files and (b) implements a programming language which can apply textual transformations to such files. It reads input files character by character (sequentially), applying the operation which has been specified via the command line (or a "chomski script"), and then outputs the line. It was developed from 2006 as a Unix and Windows utility, and is available today for Windows and Linux systems. Chomski has derived a number of ideas and syntax elements from Sed a command line text stream editor.

Usage

The following example shows a typical use of chomski, where the "-s" option indicates that the chomski expression follows: cat inputFileName | chomski -s '/(/ { until ")"; print; } clear;' > outputFileNameIn the above script, only text within brackets would be saved in the output file.

Under Unix (and Windows), chomski can be used as a filter in a pipeline:le generate_data | chomski -s '/x/{clear;add "y";}print;clear;'That is, generate the data, and then make the small change of replacing "x" with "y".

Several commands can be put together in a file called, for example, "substitute.chom"and then be applied using the "-f" option to read the commands from the file: cat inputFileName | chomski -f substitute.chom > outputFileName

Besides substitution, other forms of simple processing are possible. For example, the following uses the "plus" and "count" commands to count the number of lines in a file: cat inputFileName | chomski -s ' [-n] {plus;} <>{count;print;}'

This example used some of the following metacharacters and language features:
* The square Brackets ( [] ) indicate the matching of a character class.
* The -n' string matches a newline character.
* The <> string matches the end of the input stream (text file).
* The curly braces ({}) follow tests and group multiple statements.
* The semi-colon (;) terminates all statements,

Complex chomski constructs are possible, allowing it to serve as a simple, but highly specialised, programming language. Chomski has only one flow control statement(apart from the test structures <>, [] , // etc), namelythe "check" command, which jumps back to the @@ label (no other labels are permitted).

History

The idea for chomski arose from the limitations of regular expression engines which use a "line by line" paradigm, and the limitations on parsing nested text patterns with regular expressions.chomski evolved as a natural progression from the grep and sed command. Developmentbegan approximately in 2006 and continued sporadically.Developers (m.j.bishop) personal recollection]

Samples

In the following example, chomski adds 3 whitespace characters at the beginning of each line of input

[-n] { add ' '; } print; clear;

* ( [-n] ) match a newline character
* ({}) only execute brace blocks if the match returned true
* ( [http://bumble.sourceforge.net/machine/doc/print.txt.html print] ) print the current contents of the workspace to standard output.
* (clear) delete the contents of the workspace

The following script places brackets around urls in text [-a] { while !' '; { put; clear; add'('; get; add')'; } } print; clear;

The test structures

Conditioned commands are possible using the "test structures":Test if the workspace matches exactly the string. (This is not a regular expression match) /somestring/ {commandlist;}Test if the workspace begins with the string {commandlist;}Test if the workspace ends with the string (somestring) {commandlist;}Test if the workspace matches the character class [characterclass] {commandlist;}Test if the workspace matches any of the lines in the file =filename= {commandlist;}Test if the workspace matches the current element of the tape structure = {commandlist;}Test if the end of the input stream has been reached. (This is equivalent to the Awk END label) <> {commandlist;}

These tests may be negated by added the ! operator before (not after as in sed) the test. For example !(.) {print;}prints the workspace if it does "not" end with a period/ full-stop.

Limitations

Chomski is not a general purpose programming language. Like sed it is designed for a limited typeof usage. chomski currently does not support unicode strings since the current implementationuses standard C character arrays. Chomski does not currently have a debugger for debugging complex scripts.

ee also

*Sed
*Awk

References

External links

* [http://bumble.sourceforge.net/machine Source code and executables for chomski]
* [http://bumble.sourceforge.net/machine/doc/ Documentation for the chomski language]
* [http://sed.sourceforge.net Major sources for sed scripts, files, usage]